So how on Earth (as opposed to Wentworld) are large organizations stable?
My model, prior to reading further:
Well, what do such structures maximize? By what goals are they optimal?
The answer that seems obvious to me is “the perceived power of those at the top of such organizations”. The people constituting the-organization-as-a-whole are obviously in a suboptimal configuration for maximizing their collective economic efficiency, or their collective power. But the people in charge find themselves in nominal command of an enormous pile of resources. They are poorer, in an absolute sense, than they would’ve been if the world were in the equilibrium state in which people are arranged in more efficient economic structures. But they are relatively much more rich and influential than their underlings, which parses as optimal by their power- and status-maximizing heuristics, so it’s in their individual interests to preserve this status quo.
At the cultural level, this equilibrium is maintained by such organizations instinctively cooperating with each other to preserve it (as Zvi outlines e. g. here, I think). This is probably also culturally supported by upper managers essentially wanting to feel like “the king of their kingdom”, using historical kingdoms/governments as reference points. Said kingdom/governments, in turn, have historically ended up in big-dumb-organization configurations to maximize the power of those at their helm. (Except they spread their influence by more obvious conquest-and-pillaging, not by hiring people.)
Everyone outside mazes isn’t well-versed enough in the subject to understand what they’re signing up for when joining, or unable to imagine superior alternative configurations, or culturally indoctrinated to view such structures as inherently valuable as well. So they don’t demand larger pay, aren’t unwilling to work with large organizations, don’t vote to dissolve them, et cetera.
… so e.g. large companies or government agencies are basically runaway human monuments of dominance and submission, limited mainly by their budget.
Yeah, that seems to agree with my model.
I’d only add that this setup increases upper managers’ status outside the organizations as well, and in a fairly robust manner. It’s not quite like with the rhesus monkeys: the CEO of a Fortune 500 company, if they leave their organization and seek to join a new one, wouldn’t start at the very bottom at all. They’d often end up at the executive level as well, or even straight-up be re-hired as a CEO.
You know how the org structure is always hierarchical for some reason? And managers end up being massive communication-bottlenecks across the org, because horizontal coordination usually has to route through them? That’s not a bug, that’s a feature of an organization whose managers optimize first for dominance-status, and only secondarily for making money.
Hm, I don’t know if that’s necessarily a bug from the perspective of efficient organization design either. Inasmuch as organizations are control systems reflecting the domain they control, the hierarchy reflects the hierarchy of natural abstractions, with the managers/subsystems at higher levels “acting as natural latents”, inducing conditional independence between lower-level subsystems. This has productivity-boosting results as well: see e. g. Joel’s “development abstraction layer” argument (the people within the subsystem can focus their optimization-power solely on the domain they’re responsible for, with no distractions).
Obviously this is horribly mismanaged in practice, but it feels like a warped feature of a fundamentally productive structure.
My model, prior to reading further:
Well, what do such structures maximize? By what goals are they optimal?
The answer that seems obvious to me is “the perceived power of those at the top of such organizations”. The people constituting the-organization-as-a-whole are obviously in a suboptimal configuration for maximizing their collective economic efficiency, or their collective power. But the people in charge find themselves in nominal command of an enormous pile of resources. They are poorer, in an absolute sense, than they would’ve been if the world were in the equilibrium state in which people are arranged in more efficient economic structures. But they are relatively much more rich and influential than their underlings, which parses as optimal by their power- and status-maximizing heuristics, so it’s in their individual interests to preserve this status quo.
At the cultural level, this equilibrium is maintained by such organizations instinctively cooperating with each other to preserve it (as Zvi outlines e. g. here, I think). This is probably also culturally supported by upper managers essentially wanting to feel like “the king of their kingdom”, using historical kingdoms/governments as reference points. Said kingdom/governments, in turn, have historically ended up in big-dumb-organization configurations to maximize the power of those at their helm. (Except they spread their influence by more obvious conquest-and-pillaging, not by hiring people.)
Everyone outside mazes isn’t well-versed enough in the subject to understand what they’re signing up for when joining, or unable to imagine superior alternative configurations, or culturally indoctrinated to view such structures as inherently valuable as well. So they don’t demand larger pay, aren’t unwilling to work with large organizations, don’t vote to dissolve them, et cetera.
Yeah, that seems to agree with my model.
I’d only add that this setup increases upper managers’ status outside the organizations as well, and in a fairly robust manner. It’s not quite like with the rhesus monkeys: the CEO of a Fortune 500 company, if they leave their organization and seek to join a new one, wouldn’t start at the very bottom at all. They’d often end up at the executive level as well, or even straight-up be re-hired as a CEO.
Hm, I don’t know if that’s necessarily a bug from the perspective of efficient organization design either. Inasmuch as organizations are control systems reflecting the domain they control, the hierarchy reflects the hierarchy of natural abstractions, with the managers/subsystems at higher levels “acting as natural latents”, inducing conditional independence between lower-level subsystems. This has productivity-boosting results as well: see e. g. Joel’s “development abstraction layer” argument (the people within the subsystem can focus their optimization-power solely on the domain they’re responsible for, with no distractions).
Obviously this is horribly mismanaged in practice, but it feels like a warped feature of a fundamentally productive structure.